jujciy oj retrovirus gene therapy, I emin, 
that such non-homologous recombination does occur with retroviruses. However, the fre- 
quency of formation of highly oncogenic retroviruses is very low. It is not possible to 
quantify this frequency because there are several requirements in addition to non- 
homologous recombination to form a highly oncogenic retrovirus (Temin and Miller, 
1984). Furthermore, the structure of a minimal vector makes the probability of such 
recombination even lower than for a replication-competent retrovirus and a cellular proto- 
oncogene. 
The next question is, in the highly improbable event that an infectious virus is 
formed, what would its structure be? Theoretically, the infectious virus could contain the 
inserted gene. However, because of the packaging limit on retrovirus size and the need 
for integrity of the retrovirus genome, such a recombinant is unlikely to be packaged 
(Gelinas and Temin, 1986). Thus, any hypothetical infectious replication-competent 
recombinant would contain only some vector control sequences while the promoter, 
coding sequences, and 3’ processing sequences would derive from the cell genome. 
What would be the potential biological consequences of such a replication- 
competent retrovirus? Induction of disease(s) in the patient, spread to other people, in- 
duction of disease in other people and infection of the patient’s germ-line are some of the 
possibilities. 
It is hard to evaluate the effectiveness of such a hypothetical virus in causing 
disease. Most endogenous retroviruses are not pathogenic (Stoye and Coffin, 1985). Fur- 
thermore, we know that specific sequences in highly and weakly oncogenic retroviruses 
are responsible for their oncogenicity, and specific sequences in neurotropic retroviruses 
are responsible for neurotropism (Bishop and Varmus,1982). Although we do not know 
what sequences in lentiviruses like the virus of AIDS are responsible for disease, no 
lentivirus is known to be endogenous (Haase, 1986). Thus, induction of disease by a 
Recombinant DNA Research, Volume 12 
